Eur J Appl Physiol
DOI 10.1007/s00421-008-0773-z
ORIGINAL ARTICLE
End-tidal pressure of CO2 and exercise performance in healthy
subjects
Maurizio Bussotti Æ Damiano Magrı̀ Æ
Emanuele Previtali Æ Stefania Farina Æ
Anna Torri Æ Marco Matturri Æ Piergiuseppe Agostoni
Accepted: 12 May 2008
Ó Springer-Verlag 2008
Abstract High arterial CO2 pressure (PaCO2) measured
in athletes during exercise suggests inadequate hyperventilation. End-tidal CO2 pressure (PETCO2) is used to
estimate PaCO2. However, PETCO2 also depends on exer_ 2 ) and ventilation
cise intensity (CO2 production, VCO
efficiency (being PETCO2 function of respiratory rate). We
evaluated PETCO2 as a marker, which combines efficiency
of ventilation and performance. A total of 45 well-trained
volunteers underwent cardiopulmonary tests and were
grouped according to PETCO2 at respiratory compensation
(RC): Group 1 (PETCO2 35.1–41.5 mmHg), Group 2
(41.6–45.7) and Group 3 (45.8–62.6). At anaerobic
_
threshold, RC and peak exercise, ventilation (VE)
was
similar, but in Group 3, a greater tidal volume (Vt) and
lower respiratory rate (RR) were observed. Peak exercise
_ 2 were lowest in Group 1 and similar
workload and VO
between Group 2 and 3. Group 3 subjects also showed high
_ 2 suggesting a greater glycolytic metabolism. In
peak VCO
M. Bussotti (&) D. Magrı̀ S. Farina M. Matturri P. Agostoni
Centro Cardiologico Monzino, IRCCS, Istituto di Cardiologia,
Università degli Studi di Milano, Via Parea 4, 20138 Milan, Italy
e-mail: [email protected]
D. Magrı̀
Dipartimento di Scienza Cardiovascolari,
Respiratorie e Morfologiche, Università La Sapienza,
Rome, Italy
E. Previtali A. Torri
Istituto di Medicina Interna II, Università degli Studi di Milano,
Milan, Italy
P. Agostoni
Division of Respiratory and Critical Care Medicine,
Department of Medicine, University of Washington,
Seattle, WA, USA
conclusion, a high PETCO2 during exercise is useful in
identifying a specific respiratory pattern characterized by
high tidal volume and low respiratory rate. This respiratory
pattern may belong to subjects with potential high
performance.
Keywords
Exercise
Athletes End tidal of CO2 Ventilation Introduction
Many studies have gathered data suggesting ventilatory
limitations of aerobic exercise in athletes. Excessive alveolar to arterial O2 pressure (DPA-aO2) difference, abnormal
increases in arterial CO2 pressure (PaCO2) and haemoglobin desaturation have been demonstrated (Dempsey and
Wagner 1999; Durand et al. 2000; Rodman et al. 2002).
A high value of PaCO2 suggests an inadequate ventilation increase during exercise. This phenomenon may be
due to mechanical respiratory constraint, on reaching the
upper limit of expiratory flow rate and/or respiratory
muscle force production (Johnson et al. 1992), or to a low
chemoreceptor responsiveness (Harms and Stager 1995).
However, a lower hyperventilation and the consequent
higher PaCO2 can, per se, affect maximal exercise performance. Indeed, a reduced hyperventilation could determine
lower respiratory muscles work allowing for a lower blood
flow towards respiratory muscles and a gain of up to 10%,
in leg blood flow (Harms et al. 1997, 1998, 2000). This
mechanism delays the onset of leg fatigue and permits
greater exercise performance. Furthermore, a higher PaCO2
is associated with greater tissue and blood acidosis, which
through a rightward shift on the HbO2 saturation curve
allows greater O2 delivery to muscles.
123
Eur J Appl Physiol
End-tidal pressure of CO2 (PETCO2) is used for a noninvasive estimate of PaCO2 (Benallal and Busso 2000;
Wasserman et al. 2005). During exercise, the difference
between PaCO2 and PETCO2 is mainly related to respiratory rate (RR) because expiratory CO2 does not reach a
plateau. Consequently, for a given alveolar CO2, higher the
PETCO2 the lower is the RR. It is worthy to note that a high
RR means low ventilation efficiency, since the higher the
RR the greater is the percentage of dead space/tidal volume
ventilation. Therefore, PETCO2 derives from muscle
metabolism (amount of CO2 production), from the respiratory rate (RR) and CO2 chemoreceptor set point.
Accordingly, in normal subjects, high PETCO2 may be due
to high exercise performance and efficiency of ventilation.
Consequently PETCO2 can be proposed as a marker, which
combines performance and efficiency of ventilation.
The aim of this paper is to evaluate, in healthy welltrained subjects, if during exercise a relationship between
physical performance/efficiency of ventilation and PETCO2
exists.
_ 2 ) increase and confirmed by specific behaviour of O2
(VCO
_ VO
_ 2 ) and CO2 (VE/
_ VCO
_ 2 ) ventilatory equivalents and
(VE/
end-tidal pressure of O2 (PETO2) and PETCO2 (Beaver et al.
1986). The end of respiratory compensation (RC) was
_ VCO
_ 2 increased and PETCO2
identified when VE/
_
decreased (Beaver et al. 1986; Wasserman 1978). VE/
_
VCO2 is reported both as the slope of the relationship, and
measured from the beginning of loaded exercise to RC, and
_ VCO
_ 2 ratio.
at each exercise workload, as the actual VE/
_ 2 /DWorkRate (WR) slope was measured
The DVO
throughout the entire exercise (Wasserman et al. 2005).
We divided our study population into three groups
according to PETCO2 values at RC: Group 1 (1st tertile:
from 35.1 to 41.5 mmHg), Group 2 (2nd tertile: from 41.6
to 45.7 mmHg), Group 3 (3rd tertile: from 45.8 to
62.6 mmHg).
The investigation was approved by the local ethics
committee and subjects signed a written informed consent
before participating in the study.
Statistical analysis
Methods
A total of 45 healthy, physically well-trained volunteers
participated in the study. We defined ‘‘well-trained’’ subjects as those who had been performing aerobic exercise on
a regular basis for at least 1 year. Immediately before
exercise testing, all subjects underwent standard lung
function measurements (Vmax 29C, SensorMedics, USA).
A maximal symptom-limited cardiopulmonary exercise
test was performed on an electronically braked cycloergometer (Ergometrics-800, SensorMedics, USA), with
the subject wearing a nose clip and breathing through a
mass flow sensor (Vmax 29C, SensorMedics, USA) connected to a saliva trap. A personalized ramp exercise
protocol was chosen, aiming at a test duration of &10 min.
The exercise was preceded by 5 min of resting breath-bybreath gas exchange monitoring (rest) and by a 3 min
unloaded warm-up. A 12-lead ECG, blood pressure and
heart rate were also recorded.
Tests were evaluated by two expert readers. The
anaerobic threshold (AT) was identified by V-slope anal_ 2 ) and production of CO2
ysis of consumption of O2 (VO
Data are reported as mean ± sd. Mean values of the cardiopulmonary exercise tests are results of 20 s averages.
All data were evaluated with SPSS-PC + 13.0 statistical
software (SPSS-PC + Inc, Chicago, Illinois). We compared all cardiopulmonary data of three groups at Rest, AT,
RC and at peak exercise (Peak). The same data were also
analysed at iso-workloads (50, 100, 150 and 200 W),
where 200 W was the highest workload reached by all
subjects. All comparisons were made by one-way ANOVA
followed by a paired t-test as appropriated (Bonferroni post
hoc analysis). A P value of \0.05 was considered to
indicate statistical significance.
Results
All three groups were well matched with respect to age,
gender and BMI (Table 1). The forced expiratory volume
in the first second (FEV1) (107 ± 13, 105 ± 8 and
111 ± 9% of predicted in Group 1, 2 and 3, respectively, P = NS) and the forced vital capacity (FVC)
(113 ± 14, 113 ± 9 and 112 ± 7% of predicted in
Table 1 Demographics parameters of the study population
Groups
Gender (male/female)
P = 0.330
Age (years)
22.6 ± 2.0
P = 0.180
69.7 ± 8.5
Height (cm)
1
12/2
12/3
38 ± 8
23.1 ± 3.3
70.2 ± 10.5
174.4 ± 5.5
3
16/0
33 ± 11
24.3 ± 2.3
76.7 ± 8.2
177.7 ± 6.8
123
P = 0.275
Weight (kg)
2
P = ANOVA test result
39 ± 10
BMI (kg/m2)
P = 0.071
175.6 ± 7.9
P = 0.399
P \ 0.001 Group 2 vs. Group 1
P \ 0.001 Group 3 vs. Group 1,
Group 1, 2 and 3 respectively, P = NS) between groups
were similar.
At Rest, a significant difference between groups was
found only for PETCO2 (Table 2, Figure 1). Figure 1
_ 2 versus PETCO2 at Rest, AT,
reports the behaviour of VO
RC and Peak: Group 3 showed a significantly higher
PETCO2 at all stages of exercise. Workload, heart rate (HR)
and ventilatory parameters at all stages of exercise under
examination are reported in Table 2.
AT was reached by all subjects under similar workload
and metabolic conditions (Table 2). At this step, we
_ with, however, different ventilatory
observed similar VE
patterns between groups: indeed Vt progressively increased
and RR reduced from Group 1 to Group 3.
RC was reached at a progressively higher workload
from Group 1 to 3 (Table 2). At this stage, the difference in
_ 2 became significant.
HR, workload, Vt and VO
_ 2 and VCO
_ 2 differences
At Peak, HR, workload, Vt, VO
were maintained (Table 2), but no differences in peak HR,
_ 2 were observed between Group 2 and 3.
workload and VO
Peak exercise respiratory exchange ratio (RER) is reported
in Fig. 2; RER is higher in Group 3 with respect to the
other two groups (1.12 ± 0.11, 1.10 ± 0.07 and
1.23 ± 0.15 in Group 1,2 and 3, respectively; P = 0.005).
_ 2 , expressed as percentage of VO
_ 2 MAX predicted
Peak VO
by height, age and sex was: 113 ± 20, 132 ± 16 and
127 ± 16% in Group 1, 2 and 3, respectively (P = 0.013),
showing a lower exercise performance in Group 1 and
similar performances in Group 2 and 3.
P \ 0.05 Group 3 vs. Group 2;
P = ANOVA test result; * P \ 0.05 Group 3 vs. Group 1;
181 ± 15*
3
115.0 ± 22.5
349 ± 50
$
P \ 0.05 Group 2 vs. Group1,
48.0 ± 7.1*
39 ± 11
3.12 ± 0.51*
&
P \ 0.001 Group 3 vs. Group 2,
§
45.2 ± 5.3&
59.4 ± 10.0
P = 0.000
39.0 ± 3.3
P = 0.001 36.5 ± 2.5
P = 0.014 46.7 ± 8.3
52.0 ± 6.4
48.8 ± 6.5$
P = 0.006 41.8 ± 6.6
P = 0.310 2.59 ± 0.54
2.59 ± 0.45
44 ± 8
112.6 ± 26.6
P = 0.025 104.4 ± 21.5 P = 0.455 41 ± 9
353 ± 70$
P = 0.006 298 ± 53
165 ± 14
178 ± 9$
2
168 ± 15*
3
Peak 1
43.4 ± 7.1*
82.0 ± 19.2
308 ± 55*
28 ± 7
3.00 ± 0.63
32.7 ± 18.5
50.2 ± 4.5&
P = 0.000
43.9 ± 1.4§
P = 0.229 38.8 ± 1.9
P = 0.036 35.4 ± 6.4
27.8 ± 4.8
42.7 ± 7.0
P = 0.022 37.3 ± 5.9
P = 0.190 2.58 ± 0.50
2.47 ± 0.48
31 ± 5
82.8 ± 18.8 P = 0.470 33 ± 7
75.2 ± 16.8
P = 0.022 254 ± 39
154 ± 15
165 ± 11
1
2
RC
138 ± 19
3
P = 0.027
299 ± 64
2.38 ± 0.71
49.1 ± 11.8
199 ± 53
22 ± 5*
31.3 ± 6.7
19.4 ± 10.1
49.6 ± 3.8&
P = 0.000
P = 0.403 40.1 ± 2.3
43.9 ± 1.9$
18.2 ± 3.3
P = 0.304 21.6 ± 5.4
P = 0.045 28.2 ± 7.8
32.2 ± 7.2
1.96 ± 0.40
P = 0.042 1.89 ± 0.53
25 ± 5
50.3 ± 13.1 P = 0.932 28 ± 8
P = 0.163 171 ± 50
129 ± 14
141 ± 14
1
2
AT
67 ± 5
3
P = 0.154
48.5 ± 13.0
5.3 ± 1.4
0.88 ± 0.21
15 ± 2
13.1 ± 4.2
–
210 ± 57
4.7 ± 1.6
37.6 ± 3.1*
P = 0.005
35.6 ± 2.4
P = 0.153 33.8 ± 3.4
4.7 ± 1.9
3.8 ± 0.5
P = 0.464
4.8 ± 0.7
5.1 ± 1.1
P = 0.431 0.88 ± 0.49
0.70 ± 0.12
16 ± 3
–
P = 0.104 16 ± 5
12.7 ± 3.8
10.6 ± 1.1
P = 0.140 –
68 ± 4
72 ± 6
1
2
Rest
RR (breath/min)
_ (l/min)
VE
Work (Watt)
Groups HR (bpm)
Table 2 Cardiopulmonary data at metabolic steps in the three study groups divided according to PETCO2 values
Vt (l)
P = 0.184
_ 2 /kg (ml/kg/min)
VO
_ 2 /kg (ml/kg/min)
VCO
Pet CO2 (mmHg)
Eur J Appl Physiol
_ 2 ) and end-tidal pressure of CO2
Fig. 1 O2 consumption (VO
(PETCO2) behaviour at Rest, at anaerobic threshold (AT), at end of
respiratory compensation (RC) and at Peak in the three study groups.
_ 2 and PETCO2 between
The standard deviation and ANOVA for VO
groups are given in the table inset. Continuous line, long dashed line
and short dashed line identify respectively Group 1, 2 and 3.
_ 2 values of Group 1.
_ 2 values of Group 3 vs. VO
*P \ 0.05 VO
$
_ 2 values of Group 2 vs. VO
_ 2 values of Group 1
P \ 0.05 VO
123
123
26.6 ± 5.5
28.7 ± 5.4
P \ 0.001 Group 3 vs.
&
P \ 0.05 Group 3 vs. Group 2;
25 ± 5
22 ± 6
48.3 ± 5.4
49.3 ± 7.4
2
3
P = ANOVA test result; * P \ 0.05 Group 3 vs. Group 1;
Group 2, $ P \ 0.05 Group 2 vs. Group 1
1.97 ± 0.30
2.31 ± 0.55
P \ 0.001 Group 3 vs. Group 1;
31.0 ± 3.7
31.0 ± 4.3
§
P \ 0.001 Group 2 vs. Group 1,
26.5 ± 1.1
22.7 ± 2.2&
P = 0.000
$
29.1 ± 2.2
30.2 ± 5.9
30.9 ± 4.0
2.04 ± 0.33
30 ± 6
§
60.2 ± 9.7
1
200
P = 0.000
P = 0.000
20 ± 5*
36.0 ± 5.9
3
P = 0.002
1.90 ± 0.41
P = 0.060
23.4 ± 3.7
P = 0.996
20.4 ± 6.7
P = 0.224
23.5 ± 1.8&
P = 0.000
27.5 ± 2.2
29.4 ± 2.9
22.1 ± 4.7
19.1 ± 3.6
25.0 ± 4.1
24.5 ± 3.6
1.69 ± 0.29
1.63 ± 0.27
35.9 ± 3.3§
27 ± 8
23 ± 4
44.4 ± 7.4
2
150
P = 0.015
1
18 ± 4*
26.5 ± 3.1*
3
P = 0.004
1.60 ± 0.62
P = 0.079
16.8 ± 2.7
P = 0.457
13.7 ± 2.7
P = 0.179
25.7 ± 2.6&
P = 0.000
30.7 ± 2.3
29.2 ± 2.0
13.5 ± 3.0
14.6 ± 2.7
17.9 ± 2.2
18.7 ± 2.6
1.47 ± 0.43
1.40 ± 0.28
20 ± 5
27.0 ± 3.6
2
23 ± 5
30.9 ± 5.9
1
100
20 ± 5
P = 0.296
P = 0.033
1.33 ± 0.35
16 ± 3*
20.4 ± 4.2
3
P = 0.525
10.6 ± 1.3
P = 0.128
9.4 ± 1.7
P = 0.534
28.6 ± 3.5
34.4 ± 3.7
32.2 ± 2.7
P = 0.331
9.0 ± 2.0
8.5 ± 1.5
P = 0.334
11.2 ± 1.5
11.3 ± 1.6
1.08 ± 0.38
1.09 ± 0.33
18.9 ± 3.8
2
P = 0.032
19 ± 4
21.5 ± 5.4
1
50
Vt (l)
RR (breath/min)
_ (l/min)
VE
The main finding of our study is that in a group of healthy
physically well-trained subjects, those with the lowest
values of PETCO2 during exercise have a low exercise
_ 2
performance, as demonstrated by lower workload and VO
reached. Interestingly these subjects showed a specific
_ due to a high RR.
ventilatory pattern, featuring high VE
Group 2 and 3 subjects have the same exercise capacity,
but subjects with the highest PETCO2 have also the highest
_ 2 at peak exercise.
VCO
Groups
Discussion
Work (Watt)
Data at iso-workloads are reported in Table 3. Differences in ventilatory pattern were observed at all stages. To
evaluate the respiratory pattern throughout the exercise we
also averaged all data obtained every 50 W (from 50 to 200
_ was
where data from all subjects were obtained). Mean VE
$
*
39.3 ± 16.3, 32.5 ± 11.7 and 33.0 ± 12.2 l/min, mean
Vt was 1.56 ± 0.46, 1.54 ± 0.47 and 1.78 ± 0.61* l/min,
mean RR was 25.1 ± 7.0, 21.7 ± 5.0$ and 19.0 ± 5.2
bpm in Group 1, 2 and 3 respectively.
_ VCO
_ 2 slope was 30.1 ± 2.8, 25.4 ± 2.3§ and
The VE/
22.0 ± 3.6
in Group 1, 2 and 3, respectively
_ 2 /DWorkRate slope was 9.1 ±
(P \ 0.0001). The DVO
0.9, 9.3 ± 1.8 and 9.5 ± 1.0 ml/W/min in Group 1, 2
and 3, respectively (P = 0.614) (* P \ 0.05 Group 3 vs.
Group 1; $ P \ 0.05 Group 2 vs. Group 1; P \ 0.05
Group 3 vs. Group 2; P \ 0.001 Group 3 vs. Group 1;
P \ 0.05 Group 3 vs. Group 2; § P \ 0.001 Group 2
vs. Group 1).
Table 3 Iso-watt cardiopulmonary data in the three study groups divided according to PETCO2 values
Fig. 2 Respiratory exchange ratio (RER) values and standard
deviation at peak exercise in the three study groups
P = 0.099
_ 2 /kg (ml/kg/min)
VO
_ 2 /kg (ml/kg/min)
VCO
_ VCO
_ 2
VE/
P = 0.000
Eur J Appl Physiol
Eur J Appl Physiol
_ 2
We evaluated if the algorithm used to measure VCO
could be affected by the respiratory pattern and there_ 2
fore our results could be influenced by the VCO
_ 2 in the Sensor Medics Vmax
calculation technique. VCO
29C is measured according to the following formula:
_ 2 = STP 9 (FeCO2 - FiCO2) 9 RR/(1 - FeCO2 +
VCO
FiCO2/RER). The RR appears in the numerator of the
formula so that subjects with higher RR (Group 2) should
_ 2 . However, our
have, if anything, a higher measured VCO
results were the opposite, so that the algorithm used was
_ 2 differences.
not the cause of the observed VCO
The decision to subdivide our study population into
three different groups, using PETCO2 values obtained at
RC, was taken to evaluate the role of ventilatory patterns
on exercise performance. Indeed, RC is the exercise stage
featuring the highest PETCO2 and is also the exercise stage
less influenced by the volitional component of exercise
ventilatory regulation and exercise performance (Wasserman et al. 2005). It should be stressed that PETCO2 is a
marker, which combines exercise performance and efficiency of ventilation and by no means can be considered an
independent physiological variable (see ‘‘Introduction’’).
PETCO2 is used as a non-invasive measure of PaCO2
(Johnson et al. 1992; Wasserman 1978) in subjects with no
evidence of cardiac and lung diseases and is most useful
and indicative of PaCO2 when phase 3 of expiration is
virtually flat, a situation that may not pertain to heavy
exercise (Jones et al. 1979; Wasserman 1978). A high
PaCO2 is considered a sign of inadequate hyperventilation
(Dempsey and Wagner 1999; Martin et al. 1979) and of
exercise limitation due to the respiratory system if a subject
is near his maximal voluntary ventilation. However, and
apparently in contradiction with the previous statement,
subjects belonging to the lowest tertile of PETCO2 and
therefore to the lowest value of PaCO2, achieved the lowest
_ 2 during exercise.
peak workload and VO
Subjects with the lowest PETCO2, and probably lowest
PaCO2, are likely to have the greatest exercise-induced
acidosis, which explains the reduced exercise capacity.
Why Group 2 and 3 have a different ventilatory pattern is
much less clear but, in our opinion, very interesting. Group
_ 2
2 and 3 reached the same exercise performance (same VO
and workload) but in Group 3, PETCO2, by definition, and
_ 2 , unexpectedly, were both higher at peak exercise.
VCO
Several explanations for these results are possible. The
ventilatory pattern of subjects with higher PETCO2 at RC is
peculiar and featured high Vt and low RR throughout
_ was registered between Group
exercise, albeit a similar VE
2 and 3. This ventilatory pattern, well described in endurance athletes, offers less expenditure of ventilation in dead
space and, therefore, minor work of respiratory muscles
(Clark et al. 1983; Johnson et al. 1992). This pattern could
be related to a reduced chemoreceptor sensitivity
associated with a lower dyspnoea feeling (Rodman et al.
2002; Takano et al. 1997). Harms et al. showed that minor
tiredness in respiratory muscles, which can be experimentally obtained by mechanical unloading of these
_ 2
muscles, allows for a reduction of respiratory muscles VO
and blood flow and that this phenomenon is associated with
more than 10% leg blood flow increase (Harms et al. 1995;
Harms et al. 1997). Moreover, the same Authors suggested
that the reduction of respiratory muscles workload and the
increase in peripheral muscles flow, delays the feeling of
dyspnoea, allowing a more advanced exercise load (Harms
et al. 1998).
According to this theory, Group 3 subjects should have
registered a greater exercise performance than Group 1 and
2, but we observed that Group 2 and 3 have the same
exercise capacity. We assume that Group 3 subjects were
less fit subjects compared to those of Group 2. Indeed,
subjects in the third tertile of PETCO2 had greater RER and
_ 2 , suggesting an increased glycolytic metabolism
peak VCO
during exercise. Moreover, albeit not statistically different,
anaerobic threshold showed a trend of lower work rates in
Group 3 vs. Group 2. However, our hypothesis needs to be
verified by a study on the effects of respiratory pattern as a
guide to exercise training.
Our hypothesis that subjects with high PETCO2 present
a reduced ventilation response to exercise-induced CO2
accumulation is strengthened by the iso-watt analysis of
ventilatory pattern during exercise. Indeed, starting from
100 W, subjects with higher PETCO2 at RC showed lower
_ and of RR compared to Group 1 and a trend
values of VE
toward a higher Vt and lower RR compared to Group 2.
_ of each of the
Furthermore, if all data, all RR, Vt and VE
iso-watt analysis are averaged, the ventilatory pattern
difference between groups is clear, with lower RR and
_ in
higher Vt in Group 3, even if there is a similar VE
Group 2 and 3.
A few study limitations should be recognized. Firstly,
our research suffers an absence of blood gas analysis and
plasmatic lactate concentration values. Indeed, differences
between blood and end-tidal CO2 data are likely not to be
the same in comparing slow and fast breathing subjects.
Blood gas data, which certainly would have been useful to
support our findings, were not collected. However, Wasserman et al. (1978) reported a difference between PaCO2
and PETCO2 values of approximately +2.5 mmHg at rest
to -4 mmHg during heavy work, which is below the
difference in PETCO2 that we observed. Moreover, Jones
et al. (1979) formulated an equation to predict PaCO2
directly from PETCO2 values. All these works support a
good reliability of PETCO2 as an indirect measure of
PaCO2 in healthy subjects (Benallal and Busso 2000).
However, our study was aimed at investigating the relationship between physical performance and PETCO2
123
Eur J Appl Physiol
considered as a marker, which combines efficiency of
ventilation and exercise performance and not simply as an
indicator of PaCO2.
Secondly, we used a cycloergometer. Therefore, we do
not know if the observed respiratory pattern is also present
in exercise performed with other ergometers e.g. a
treadmill.
Thirdly, the cardiopulmonary tests were performed in
our laboratory randomly during the year, without taking the
subjects’ training period into consideration. This could be
an important point of remark, because training modifies the
respiratory pattern through an alteration of chemo and
muscular receptors responsiveness.
Fourthly, the population we analysed was a mixed
population of physically well-trained subjects who practised varied sports and were therefore not athletes in a
specific area. So our data needs to be confirmed for each
specific sport.
_ behaviour occurs through
Finally, because specific VE
the entire exercise and is present even at rest (see differences in resting PETCO2), it is possible, but totally
unproven, that in the well-trained subjects, the ventilatory
pattern pertains to regular life activity and not only to
maximal exercise performance.
In conclusion, in a healthy physically well-trained
population, PETCO2 values during exercise could be useful
in identifying particular respiratory patterns and their
underlying physiological mechanisms. This parameter,
which is easily and non-invasively detectable with a cardiopulmonary exercise test, could represent a tool in future
studies on the relationship between ventilation and exercise
performance. Indeed, it might be that subjects with the
highest PETCO2 could, with training, improve their exercise performance more than the participants in Group 2.
Consequently, our study raises more questions than it
provided answers. Further studies are certainly required to
evaluate if and how training affects the respiratory pattern
during exercise and whether PETCO2 analysis during
exercise can drive training methodologies.
Acknowledgment We are indebted to Prof. Brian Whipp for the
constructive critiques during preparation of the manuscript.
123
References
Beaver WR, Wasserman K, Whipp BJ (1986) A new method for
detecting the anaerobic threshold by gas exchange. J Appl
Physiol 60:2020–2027
Benallal H, Busso T (2000) Analysis of end-tidal and arterial PCO2
gradients using a breathing model. Eur J Appl Physiol 83(4–
5):402–408
Clark JM, Hagerman FC, Gelfand R (1983) Breathing patterns during
submaximal and maximal exercise in elite oarsmen. J Appl
Physiol 55:440–446
Dempsey JA, Wagner PD (1999) Exercise-induced arterial hypoxemia. J Appl Physiol 87:1997–2006
Durand F, Mucci P, Prefaut C (2000) Evidence for an inadequate
hyperventilation inducing arterial hypoxemia at submaximal
exercise in all highly trained endurance athletes. Med Sci Sports
Exer 32(5):926–932
Harms CA, Stager JM (1995) Low chemoresponsiveness and
inadequate hyperventilation contribute to exercise-induced hypoxemia. J Appl Physiol 79:575–580
Harms CA, Babcock MA, McClaran SR et al (1997) Respiratory
muscle work compromises leg blood flow during maximal
exercise. J Appl Physiol 82(5):1573–1583
Harms CA, Wetter TJ, McClaran SR et al (1998) Effects of
respiratory muscle work on cardiac output and its distribution
during maximal exercise. J Appl Physiol 85(2):609–618
Harms CA, Wetterb TJ, St. Croix CM et al (2000) Effects of
respiratory muscle work on exercise performance. J Appl
Physiol 89:131–138
Johnson BD, Saupe KW, Dempsey JA (1992) Mechanical constraints
on exercise hyperpnea in endurance athletes. J Appl Physiol
73:874–886
Jones NL, Robertson DG, Kane JW (1979) Difference between endtidal and arterial PCO2 in exercise. J Appl Physiol 47(5):954–
960
Martin BJ, Sparks KE, Zwillich CW et al (1979) Low exercise
ventilation in endurance athletes. Med Sci Sports 11(2):181–185
Rodman JR, Haverkamp HC, Gordon SM et al (2002) Cardiovascular
and respiratory system responses and limitations to exercise. In:
Weisman IM, Zeballos RJ (eds) Clinical exercise testing. Prog
Respir Res Karger, vol 32, Basel
Takano N, Inaishi S, Zhang Y (1997) Individual differences in
breathlessness during exercise, as related to ventilatory chemosensitivities in humans. J Physiol 499.3:843–848
Wasserman K (1978) Breathing during exercise. Physiology in
Medicine series. N Engl J Med 298:780–785
Wasserman K, Hansen JE, Sue DY et al. (2005) Normal values.
Arterial and end-tidal carbon dioxide tensions. In: Wasserman K,
Hansen JE, Sue DY et al (eds) Principles of exercise testing and
interpretation, 4th edn. Lippincott Williams & Wilkins,
Philadelphia